HAL Id: hal-00897824https://hal.archives-ouvertes.fr/hal-00897824

Submitted on 1 Jan 1981

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

Digestive enzymes in the germ-free animalT. Corring, Catherine Juste, C. Simoes-Nunes

To cite this version:T. Corring, Catherine Juste, C. Simoes-Nunes. Digestive enzymes in the germ-free animal. Repro-duction Nutrition Développement, 1981, 21 (3), pp.355-370. �hal-00897824�

Digestive enzymes in the germ-free animal

T. CORRING, Catherine JUSTE, C. SIMOES-NUNES

Laboratoire de Physiologie de la Nutrition, INRA,78350 Jouy en Josas, France.

Summary. The digestive physiology of the germ-free animal has a number of characte-ristics (cecal hypertrophy, slower small intestine cell renewal, slower gastric emptying andintestinal transit) which distinguish it from that of the conventional animal. If the germ-freemodel is to be used to determine the role of gastrointestinal microflora in the nutrition of theconventional animal, it is essential to complete the study of these characteristics by data ondigestive enzymes in the germ-free. The present paper analyzes these data.

There is little information on salivary amylase and none on gastric proteolytic enzymesand intestinal peptidases. More complete data on exocrine pancreas enzymes and intestinaldisaccharidases show that the digestive equipment is similar in germ-free and conventionalanimals. Bile salts, not considered as digestive enzymes, are qualitatively and quantitativelydifferent, depending on the digestive tract bacterial environment.

In general, the germ-free animal has some characteristics which should permit betterutilization of the diet ingested. Measurements of apparent digestibility do not confirm thishypothesis since results obtained in germ-free and conventional animals of the same speciesare contradictory.

The germ-free animal is certainly a very useful tool for studying the role of

gastrointestinal microflora in the nutrition of its host. The large intestine contains thelargest bacterial population (1 to 3.1010 bacteria/g fresh contents) and the greatestnumber of microbial species in the gut (Ducluzeau and Raibaud, 1975 ; Schaedler,1973). In some cases, bacteria may efficiently intervene in the digestive utilization ofthe diet ingested by the animal (R6rat, 1978).

However, a number of experiments in the germ-free animal have shown that it

cannot be considered simply as a conventional animal, deprived of gastrointestinalmicroflora, since certain characteristics of its digestive physiology distinguish it

from the conventional animal (see reviews by Gordon et al.,1966 ; Combe et al., 1976).A knowledge of these characteristics is essential when determining the part due tobacteria in the digestion of an ingested diet.

The germ-free state in rodents and lagormorphs leads to a substantial increasein cecal sac contents and size (Glimstedt, 1936 ; Wostmann and Bruckner-Kardoss,1959 ; Wostmann, Bruckner-Kardoss and Knight, 1968 ; Pleasants, 1959). The sameobservation was made in the rat (Lee and Moinuddin, 1958), mouse (Meynell, 1963)and guinea-pig (Jervis and Biggers, 1964) after antibiotic treatment.

The morphology of the small intestine in the germ-free rat (Gordon and Wost-mann,1960), guinea-pig (Sprinz et al.,1961) and chicken (Thorbecke et al.,1957) differsby a lower weight, a smoother mucosa and a smaller reticulo-endothelial cell popu-lation. The small intestine mucosal surface area of the germ-free rat is on an average30 p. 100 lower than that of the conventional animal ; this reduction is marked in themid and lower parts of the small intestine and relatively slight in the upper segments(Gordon and Bruckner-Kardoss, 1961). But, according to Meslin (1971), the mucosalsurface area of the proximal intestine is larger and that of the distal part more reducedin the germ-free than in the conventional rat. After deduction of cecal content weight,there was no difference between germ-free and conventional rats when consideringthe total surface of the intestinal mucosa relative to body weight (Meslin, Sacquetand Guenet, 1973).

Some differences between germ-free and conventional states were also empha-sized in the epithelial renewal rate of the small intestine. The transit time of cells

moving from the crypts to the villus tips is twice as long in the ileum of the germ-freemouse as in that of its conventional counterpart and depends on cell production which,in turn, is related to the number of cells in the proliferative pool and the duration ofthe generative cycle (Abrams, Bauer and Sprinz, 1963). The generative cycle time forduodenal crypt cells is longer in the germ-free than in the conventional mouse

(13.6 h vs 11.2 h) ; in the former, the generation rate of proliferative cells is reduced

by 20 p. 100 and the size of the proliferative cell population is decreased by about37 p. 100 (Lesher, Walburg and Sacher, 1964).

The transit rate of the intestinal cells in the germ-free rat is reduced by about30 p. 100 and the mitotic index of the Lieberkuhn crypts is lower (Guenet et al., 1970).In the chicken, the rate of epithelial cell migration is lower in the germ-free than in theconventional state, but the relative differences are greater in the lower than in the

upper intestine (Rolls, Turvey and Coates, 1978).Gastric emptying and small intestinal transit of the ingested food were found to

be slower in the germ-free than in the conventional state. Six hours after a labelled

test-meal, the cecum of the germ-free mouse retained a greater percentage of radioac-tivity than the cecum of the conventional animal, and the passage of the marker intothe feces was slower (Abrams and Bishop, 1967). Gastrointestinal transit was alsoslower in the germ-free than in the conventional rat (Sacquet, Garnier and Raibaud,1970 ; Riottot et al., 1980), and tended to be longer in germ-free rats fed irradiateddiets than in those fed autoclaved diets (Riottot et al.,1980).

The aim of the present review is to complete our knowledge of the characteristicsof germ-free animal digestive physiology by analyzing data on digestive enzymes.These enzymes (proteolytic enzymes acting on proteins, glycolytic enzymes on carbo-hydrates and lipolytic enzymes on lipids) will be discussed in order according to thenature of the dietary substrate they hydrolyze.

Proteolytic enzymes

The first step in the enzymatic hydrolysis of dietary proteins occurs in the gastriclumen under the action of pepsins and HCI (Seijffers, Segal and Miller, 1963). Hydro-

lysis then continues in the intestine where proteins and peptides from gastric hydrolysisare submitted to the action of pancreatic proteolytic enzymes and intestinal peptidases(Gray and Cooper, 1971 ; Heizer, Kerley and lsselbacher, 1972).

Gastric enzymes. - Proteolytic enzyme hydrolysis in the stomach is mainly per-formed by the pepsins secreted in an inactive form (pepsinogens) and activated byHCI. To our knowledge, no data are available on pepsinogen biosynthesis and secre-tion in germ-free animals. Roze et al. (1977) reported that basic HCI secretion washigher (X 2.7) in the germ-free than in the conventional rat and that basal gastrine-mia in these animals was not different.

The stomach of the germ-free guinea-pig has the same aspect as that of the con-ventional animal (Philips, Wolfe and Gordon, 1959), and there is no histological dif-ference in the gastric mucosae of germ-free and conventional mice (Savage, Schae-dler and Dubos, 1967). However, these results were obtained in experiments dealingwith a microscopic study of gland anatomy, and they do not provide any informationon the cellular aspect in either environment.

Pancreatic enzymes. - Pancreatic proteolytic enzymes (chymotrypsinogen, tryp-sinogen, procarboxypeptidases A and B, proelastase) are secreted, in a inactive form,by the exocrine pancreatic cells and are activated in the intestinal lumen by entero-kinase, an intestinal enzyme.

The tissue ’levels of chymotrypsinogen and trypsinogen appear to be similar inboth germ-free and conventional states (table 1) in the chicken (Coates, Hewitt andGolob, 1970) and the rat (Lepkovsky et al., 1966 ; Reddy, Pleasants and Wostmann,1969).

In addition, there is no difference between the germ-free and conventional chicken(Lepkovsky et al., 1964) and rat (Lepkovsky et al., 1966 ; Reddy, Pleasants and Wost-mann, 1969 ; Genell, Gustafsson and Ohlsson, 1976) as regards pancreatic proteolyticactivity in the small intestine contents (table 1). According to Malis et al. (1974, 1976),chymotrypsin activity is higher in the ileal contents of the germ-free than of the conven-tional rabbit (table 1).

According to Borgstrbm et al. (1959) and Loesche (1968), pancreatic proteolyticenzymes are largely inactivated by the intestinal microflora in the distal part of thedigestive tract. The concentrations of pancreatic proteases are lower in the cecal con-tents of the conventional chicken (Lepkovsky et al., 1964), rat (Loesche, 1968 ; Reddy,Pleasants and Wostmann, 1969), mouse (Loesche, 1968) and rabbit (Malis et al., 1974,1976) than in those of their germ-free counterparts (table 2). Chymotrypsin and elas-tase activities appear to be more sensitive to bacterial inactivation than is trypsinactivity (Malis et al., 1976 ; Genell, Gustafsson and Ohlsson, 1976).

Moreover, data obtained by Genell, Gustafsson and Ohlsson (1976) do not indicateany enzyme reabsorption in the large intestine of the germ-free animal since bothtrypsin and elastase appeared in the feces in the same concentrations as in the cecalcontents.

The absence of microflora in the rat does not cause any modifications in pancrea-tic weight or histological structure (Reddy, Pleasants and Wostmann, 1969). Theatrophy and fibroticchan9es observed in the germ-free rat by Geever, Daft and Leven-son (1965) might be explained by the presence of tween 80 and deficiency or lack of

selenium and chromium in the diet used by the authors (Reddy, Pleasants and Wost-mann, 1969). In addition, the trypsic inhibitor in the diet induced the same pancreaticweight increase in the chicken (Coates, Hewitt and Golob, 1970) and the rat (Kwonget al., 1971), whatever the bacterial environment. To our knowledge, no data areavailable on intestinal peptidases in the germ-free animal.

Glycolytic enzymes

Dietary carbohydrates are hydrolyzed by salivary and mainly pancreatic amy-lases ; in the intestinal mucosa, they are hydrolyzed by disaccharidases. It must be

recalled that there is no cellulase in the intestine of germ-free animals, this enzymebeing of bacterial origin.

Salivary amylase. - Data on this enzyme are scarce. Szylit (1973) and Ivorec-Szylit, Raibaud and Schellenberg (1973) showed the existence of an amylase in the cropof germ-free fowl. The action of the enzyme may be understood by some studies sho-wing that heated starches are completely degraded in the germ-free animal beforereaching the duodenum (chicken : Masson, 1954 ; Bewa, Charlet-Lery and Szylit,1979 ; other monogastric species : Baker et al., 1950).

Pancreatic amylase. -There is no difference in the level of pancreatic tissue amy-lase activity (table 3) in germ-free and conventional rats (Lepkovsky et at., 1966 ;Reddy, Pleasants and Wostmann, 1969) and chickens (Coates, Hewitt and Golob,1970). According to Reddy, Pleasants and Wostmann (1969)studying the rat, Lepkovskyet al. (1964) the chicken, and Yoshida et al. (1968) the rabbit, pancreatic amylase is

partially inactivated in the distal part of the digestive tract in the conventional animal(table 3). On the contrary, Lepkovsky et al. (1966) found a higher enzyme activily in

the distal part of the intestine of the conventional rat, whereas Borgstram et at. (1959)showed no difference between the fecal amylase activities of germ-free and conven-tional rats.

Intestinal disaccharidases.―The hydrolytic products, resulting from amylaseaction on dietary carbohydrates, are submitted to the action of different disacchari-dases distributed along the small intestine. The disaccharida.se distribution patternis similar in germ-free and conventional rats (Dahlqvist, 1963 ; Dahlqvist and Thom-son, 1964 ; Reddy and Wostmann, 1966), chickens (Siddons, 1969) and rabbits (Maliset at., 1974). In addition, age changes in disaccharidase activities and the effect ofdiet on these activities are not influenced by the bacterial environment. Thus, thedecrease in lactase and cellobiase and the increase in maltase, invertase and trehalaseat weaning are similar in germ-free and conventional rats (Reddy and Wostmann,1966). The microflora does not change disaccharidase adaptation to the dietarysubstrate in the rat (Reddy, Pleasants and Wostmann, 1968), and a mash diet, comparedto a liquid diet, leads to the same stimulation of enzyme activities in both germ-freeand conventional chickens, taking into account the different weight profile of the germ-free animal (Siddons and Coates, 1972). The reduction of sucrase, maltase and lactaseactivities in the germ-free rat after 48-hr starvation is similar to that observed in the

gnotobiotic animal (Ecknauer and Raffler, 1978). Moreover, Larner and Gillepsie(1957) and Reddy and Wostmann (1966) showed that enzyme biosynthesis is not

affected by intestinal bacteria.

The data on each of the enzyme activities in both germ-free and conventionalstates are contradictory. Maltase, sucrase, palatinase and lactase activities are similar

in germ-free and conventional chickens (Siddons, 1969 ; Siddons and Coates, 1972)(table 4) ; about 97 p. 100 or more of this activity can be recorded in the intestinal wall.The germ-free rat exhibits the same glucosidase, maltase and sucrase activities as its

conventional counterpart (Larner and Gillepsie, 1957 ; Dahlqvist, Bull and Gustafsson,1965). In contrast, maltase, sucrase, trehalase, cellobiase and lactase activities are

higher in the germ-free rat after weaning (table 4) (Reddy and Wostmann, 1966 ;Kawai and Morotomi, 1978). In the same species, Dahlqvist, Bull and Gustafsson (1965)only found higher activities for lactase and cellobiase in the germ-free animal. Accor-ding to Yoshida et al. (1968), maltase, sucrase, trehalase, lactase and cellobiaseactivities are higher in the small intestine mucosa of the germ-free than of the conven-tional rabbit (table 4). Maltase, sucrase and lactase activities are higher in the smallintestine mucosa of the germ-free rat, whereas trehalase and cellobiase activities arethe same as in the conventional animal (Reddy and Wostmann, 1966). According tothe latter authors, the lower enzyme activity in conventional animals might be explain-ed in two ways, if one considers that disaccharidases are located intracellularly andthat their action cannot be exerted in the intestinal contents (Dahlqvist and Borgstr6m,1961). On the one hand, enzymes would be partially inactivated inside the mucosal cellby the bacterial products which enter the cell wall. On the other hand, the higherenzyme activity in the germ-free animal would reflect a more advanced mucosal cellmaturity and thus a more elevated enzyme level. But this eventuality is not known inall species.

The desquamated cells lead to disaccharidase activity in the intestinal contents,and the question is to determine whether the number of these cells depends on thebacterial environment. According to Meslin, Sacquet and Raibaud (1974), the produc-tion of epithelial cells by cell column is the same in the duodenum and jejunum ofgerm-free and conventional rats but higher in the ileum of the conventional animals.Different authors agree that the enzyme level in the distal part of the digestive tractof the germ-free animal is higher due to bacterial inactivation of the enzymes in theconventional animal (Borgstrdm et al., 1959 ; Dahlqvist, Bull and Gustafsson, 1965 ;Reddy and Wostmann, 1966 ; Kawai and Morotomi, 1978).

Lipolytic enzymes

Lipids are mainly hydrolyzed in the intestinal lumen through the action of pan-creatic lipase. Lipolysis needs the presence of bile salts and colipase. The latter wasonly found recently (Maylie et al., 1971) and, while a few data are available on theconventional animal, they are completely lacking on the germ-free.

Pancreatic lipase.―There is no difference between the pancreatic tissue lipaselevels in germ-free and conventional rats (Lepkovsky et al., 1966 ; Reddy, Pleasants *sand Wostmann, 1969) (table 5). In the small intestine contents of the rat, the sameauthors showed that lipase activity is not modified by the bacterial environment. In

the cecum and colon of the rat, Reddy, Pleasants and Wostmann (1969) have reportedthat enzyme activity is significantly higher in the germ-free animal ; this disagreeswith the results of Lepkovsky et al. (1966) in the rat and of Lepkovsky et al. (1964) in the

chicken, showing no difference in the distal part of the digestive tract, whatever thebacterial environment (table 5).

According to Reddy, Pleasants and Wostmann (1969), the stability of pancreaticlipase is intermediate between that of pancreatic proteases and pancreatic amylase.

Bile salts. -The greatest number of studies in the germ-free animal concern bileand bile salts. In conventional animals, bile acids are excreted in a conjugated formin the intestinal lumen where they are submitted to microflora action. Bacterial hydro-lases can split the bond between bile acid and taurine or glycine (Aries and Hill,1970a) and bacterial dehydrogenases can either form ketones or reduce ketones toalcohols (Aries and Hill, 1970b). Cholic acid is transformed into deoxycholic acidand chenodeoxycholic acid into lithocholic acid under the action of bacterial 7a-dehy-droxylase (Aries and Hill, 1970b). This 7«-dehydroxylase does not act on conjugatedbile salts or on their methyl esters (Aries and Hill, 1970b). Some bacteria can form a5«-cholanic acid from bile acids with a ketone function in position 3 after induction ofa double bond in position 4 (Aries and Hill, 1970c). Bile acids are reabsorbed viapassive diffusion in all regions of the intestine and via active absorption in the ileum(Schiff, Small and Dietschy, 1972).

The absence of microflora in the germ-free leads to reabsorption of essentiallyprimary bile salts, i.e. salts which have not been modified. In addition, the bile salt

composition of the secreted bile is simpler than that of the conventional animal (Sac-quet, 1971 ; Demarne, Sacquet and Garnier, 1972). In the germ-free rabbit (Hofmann,Mosbach and Sweeley, 1964), 95 p. 100 of the bile salts are constituted of cholic acid,whereas in its conventional counterpart, 95 p. 100 contain deoxycholic acid, reflectingthe absence in the rabbit liver of any enzyme responsible for rehydroxylation ofdeoxycholic acid in position 7. In the germ-free rat, about 98 p. 100 of the bile salts inthe bile are cholic and p-muricholic acids (Wostmann, 1973), and hyodeoxycholic,deoxycholic and lithocholic acids are absent in that secretion (Demarne, Sacquet andGarnier, 1972). The conventional rat, fed raw potato starch or lactose, secreted bileacids qualitatively similar to those found in the germ-free animal, indicating that verylittle microbial dehydroxylation occurred with some diets (Kellogg, 1971).

The bile acid pool is larger and fecal bile acid excretion is lower in the germ-freerat (Kellogg and Wostmann, 1969a ; Kellogg, 1971 ; Sacquet et al., 1975) and thegerm-free mouse (Eyssen, Parmentier and Mertens, 1976) (table 6). In the germ-freerat harbouring bacteria that reduce cecal distension but do not metabolize bile acids,the values of the intestinal pool and of fecal bile acid excretion are located betweenthose of germ-free and conventional animals (Sacquet et al., 1976). According toWostmann et al. (1976), the addition of lactose to the diet results in different bile saltcompositions in germ-free and conventional rats ; bile acid metabolism is not affected,contrary to data reported by Sacquet, Leprince and Riottot (1979). In addition, thelatter authors and Riottot et al. (1980) showed that the intestinal pool and fecal bileacid excretion mainly depend on small intestine transit time which is longer in thegerm-free than in the conventional rat.

In the germ-free state, a higher intestinal bile acid pool would determine a greaterbile acid flow through the liver, leading to a decrease in biosynthesis mediated by afeedback mechanism (Sacquet, Leprince and Riottot, 1979). Moreover, the more activeznterohepatic circulation of bile salts in the germ-free animal leads to a decrease in thecatabolism of hepatic cholesterol (Kellogg, 1971 ; Wostmann, 1973). Liver cholesterolconcentrations are three times higher in the germ-free rat fed a purified diet containing0.5 p. 100 cholesterol than in the conventional animal (Kellogg and Wostmann,1969b).

According to Einarsson, Gustafsson and Gustafsson (1973), 7a-hydroxylase acti-vity is higher in the conventional state, and the effect of bacteria may be partly mediatedvia increased cholesterol 7«-hydroxylation in the liver since the activity of this enzymeis supposed to be the rate-limiting step in the conversion of cholesterol into bile acids(Danielssson, Einarsson and Johansson, 1967). However, the data obtained by Ukai!Tomura and Ito (1976), showing enlarged pools of both cholesterol and bile acids inthe enterohepatic circulation of the germ-free rat, suggest that the decrease in hepaticcholesterol catabolism might be due to slower total metabolism in the germ-free ani-mal.

Other intestinal enzymes

No difference was found between germ-free and conventional rats as to the pat-terns of hexokinase, glucose-6-phosphatase, pyruvate kinase and lactate dehydroge-nase distribution along the gastrointestinal tract (Kawai and Morotomi, 1978). Neitherdid these authors find any difference in hexokinase or pyruvate kinase activities in

these animals, whereas glucose-6-phosphatose activity in the whole digestive tract andlactate dehydrogenase activity in the cecum were higher in the germ-free rats.

Histochemical studies in the mouse do not show any significant differences bet-ween germ-free, monocontaminated and conventional animals as to the duodenaland jejunal enzyme activity of NAD and NADP succinic dehydrogenase and esterase(Herskovic et al., 1967). The activity of alkaline-phosphatase and adenosine-triphos-phatase, which are involved in active intestinal transport systems, is higher in the germ-free mouse (Jervis and Biggers, 1964) and rat (Reddy, 1971 ; Kawai and Morotomi,1978) than in their conventional counterparts. Acid phosphatase, involved in local

defense mechanisms, has a lower activity in the germ-free than in the conventionalmouse (Jervis and Biggers, 1964).

Conclusions.

Some general conclusions may be drawn from the above analysis. Firstly, data ondigestive enzymes do not actually provide a complete picture of digestive enzymeequipment in the germ-free animal. Indeed, information on salivary amylase is

scarce, and we have no data on gastric proteolytic enzymes and intestinal peptidases.Secondly, taking into account what we know, i.e. pancreatic enzymes and intes-

tinal disaccharidases, it can be concluded that the digestive enzyme equipment of thegerm-free is similar to that described in its conventional counterpart. Even if they arenot directly related to enzymes, the bile salts secreted in bile are qualitatively andquantitatively different in germ-free and conventional states. Nevertheless, it is not

clear if such differences could affect the process of dietary lipid hydrolysis by pancrea-tic lipase in the germ-free.

Finally, most authors agree that enzyme activity in the distal part of the digestivetract is higher in the germ-free than in the conventional animal. In the latter, endoge-nous enzyme secretions are inactivated by bacteria at that level, indicating the presenceof a greater amount of endogenous matter in the intestinal lumen of the germ-freeanimal, as shown by some authors. Combe et al. (1965), Reddy, Pleasants and Wost-mann (1969), Combe, Arnal and Sacquet (1967) and Loesche (1968) studying the rat,Loesche (1968) and Whitt and Demoss (1975) the mouse, and Salter and Coates (1971)the chicken, reported a large accumulation of endogenous proteins and free aminoacids in the germ-free state. The nature of the nitrogen present in the cecum differs ingerm-free and conventional animals. In the former, it is mainly represented by endo-genous proteins : enzymes, mucoproteins, desquamated cells, more of less long-chain peptides and free amino acids. In the conventional animal, the cecal nitrogenousmaterial (dietary residues, microorganisms and lesser amounts of endogenous pro-teins) is primarily undigested and unabsorbed. The accumulation of nitrogenous mate-rial and water in the cecum of germ-free animals is in part due to the fact that their

cecal absorption is inhibited (Combe and Gordon, 1969 ; Loeschke and Gordon,1970), whereas the excretion of plasma proteins is similar in germ-free and conven-tional rats (Levenson, Gruber and Kan, 1969). Accumulation of carbohydrates wasalso reported in the cecum of germ-free animals. The cecal contents of germ-freerats and mice are higher (20 to 200 p. 100) in soluble carbohydrates than those ofconventional animals. The soluble carbohydrate level is 150 p. 100 higher in the fowlcecum 8. hrs after the beginning of the meal (Szylit, 1971). This cannot be explained bya decrease in amylase activity in the germ-free, but rather by the fact that starch is

degraded more rapidly when the grains are surrounded by live bacteria (Baker et al.,1950). Although it is beyond the scope of this review, a general look at the processesof digestion in the germ-free animal is informative. Three main factors are involved :

(i) 9qstnpi,ptestinal transit determining the distribution of dietary substrates in the dif-ferent parts of the intestine, (ii) gastrointestinal enzymes acting on the dietary substratesand contributing to the release of products which are submitted to (iii) absorption.Gastrointestinal transit is slower in germ-free than in conventional animals ; thus, it

can be assumed that intestinal enzyme digestion is more efficient in the germ-free sinceenzyme contact with the dietary substrate is more durable and it has been shownthat enzyme activity remains high in the distal gut contents. The absorption of enzymehydrolytic products is generally higher in the germ-free than in the conventional ani-mal. According to Whitt and Demoss (1975), the lower free amino acid concentrationsin the distal part of the germ-free small intestine reflects the greater absorption rate ofthose products. The absorption of 14C -L-methionine is twofold higher in the germ-freethan in the conventional mouse (Herskovic et al., 1967), whereas Riedel, Scharrer andL6sch (1972) showed very little difference in the intestinal absorption of 14C-leucinein germ-free and conventional fowl. Xylose absorption is twice as quick in the germ-free rat and mouse as in their conventional counterparts (Heneghan, 1963). Lipidabsorption depends on the type of fat ingested ; saturated fatty acids, such as palmiticand stearic acids, are better absorbed in the germ-free than in the conventional rat(Demarne et al., 1970).

In theory, the germ-free animal possesses a whole set of physiological characte-ristics allowing a better digestive utilization of the diet ingested. The measurement ofapparent digestibility is a criterion commonly used to estimate the digestive utilizationof the diet. The results obtained on germ-free and conventional animals are contra-dictory. Apparent nitrogen digestibility may be higher (Yoshida et al., 1968 ; Yama-naka et al., 1972) or lower (Combe, 1971 ; Corring, Moreau and Ducluzeau, 1979) inthe germ-free than in the conventional animal. Coates, Hewitt and Salter (1973) didnot find any difference in the apparent nitrogen digestibilities of germ-free and conven-tional chickens. Apparent digestibility of nitrogen-free extract decreases from 80 to66 p. 100 in the rabbit without flora (Yoshida et al., 1968). In the absence of microflora,apparent lipid digestibility (Luckey, 1963 ; Yamanaka et al., 1972), particularly that oflong-chain saturated fatty acids (Boyd and Edwards, 1967 ; Demarne et al.,1970,1972),is higher. The reasons for such a discrepancy in these results have not been fullyunderstood. However, it seems likely that the microflora may play a role in the diges-tive utilization of the diet in the conventional animal. The amounts and types of diges-tive residues, coming from the small bowel and passing through the ileo-cecal valveinto the distal digestive tract, may vary according to the nature of the dietary compo-

nents (Rerat,1978). In the germ-free animal, these residues, augmented by endogenoussecretions, would be found in the feces and thus change the apparent digestibility ofthe diet. In the conventional animal, the microflora may reduce apparent nitrogendigestibility as, for instance, when fecal exogenous nitrogen excretion is low, andincrease that digestibility when the supplies of undigested amino acids are high (R6rat,1978). The experiments of Corring, Moreau and Ducluzeau (1979) confirm the latterpoint. Indeed, in conventional rats, the nitrogen digestibility of a semi-purified dietis only slightly (but significantly) reduced by suppression of pancreatic secretion. Incontrast, nitrogen digestibility is definitively decreased by pancreatic deviation in

germ-free rats. It would thus seem that the presence of microflora leads to an improve-ment of nitrogen digestibility.

Re!u en juillet 1980.

Accepte en janvier 1981.

Résumé. L’animal germ-free présente un certain nombre de caractéristiques dans saphysiologie digestive qui le différencient de son homologue conventionnel : hypertrophie ducaecum, ralentissement du renouvellement cellulaire dans l’intestin grêle, ralentissement del’évacuation gastrique et du transit intestinal. Afin de compléter l’étude de ces caractéris-tiques, essentielle si l’on veut utiliser l’animal germ-free pour préciser le rôle de la micro-flore gastrointestinale dans la nutrition du conventionnel, un ensemble de données relativesaux enzymes digestives chez le germ-free sont analysées dans cette revue.

Il apparaît que les informations sont rares en ce qui concerne l’amylase salivaire etinexistantes pour ce qui est des enzymes protéolytiques gastriques et des peptidases intesti-nales. Plus nombreuses en ce qui concerne les enzymes du pancréas exocrine et les disac-charidases intestinales, elles montrent que l’équipement enzymatique digestif de l’animalgerm-free est semblable à celui de l’animal conventionnel. Les sels biliaires, non considéréscomme enzymes digestives sont, qualitativement et quantitativement différents en fonctionde l’environnement bactérien du tube digestif.

Si l’on reprend les principaux aspects de la physiologie digestive, l’animal germ-fréeprésente un certain nombre de caractéristiques qui devraient lui permettre de mieux utiliserl’aliment ingéré. Les mesures de digestibilité apparente ne permettent pas de valider cettehypothèse puisque chez une même espèce les résultats obtenus chez l’animal germ-free etson homologue conventionnel sont contradictoires.

References

ABRAMS G. D., BAUER H., SPRINZ H., 1963. Influence of the normal flora on mucosal morphologyand cellular renewal in the ileum. A comparison of germ-free and conventional mice. Lab.Invest., 12, 355-364.

ABRAMS G. D., BISHOP J. E., 1967. Effect of the normal microflora on gastro-intestinal motility. Proc.Soc. exp. Biol. Med., 126, 301-304.

ARIES V., HILL M. J., 1970a. Degradation of steroids by intestinal bacteria. I. Deconjugation of bilesalts. Biochim. biophys. Acta, 202, 526-534.

ARIES V., HILL M. J., 1970b. Degradation of steroids by intestinal bacteria. li. Enzymes catalyzing theoxido-reduction of the 3a, 7a and 12a-hydroxy group in cholic acid and the dehydroxylationof 7-hydroxyl group. Biochim. biophys. Acta, 202, 535-543.

ARIES V., HILL M. J., 1970c. The formation of unsaturated bile acids by intestinal bacteria. Biochem. J.,119,37.

BAKER F., NASR H., MORRICE F., BRUCE J., 1950. Bacterial breakdown of structural starch productsin the digestive tract of ruminant and non-ruminant mammals. J. Path. Bact., 52, 617-638.

BEWA H., CHARLET-LERY G., SZYLIT O., 1979. R61e de la microflore digestive et de la structurecristalline de I’amidon dans la digestion et l’utilisation des regimes chez le poulet. Etude destubercules tropicaux. Ann. Nutr. Aliment., 33, 213-231.

BORGSTRZSM B., DAHLQVISTA., GUSTAFSSON B. E., LUNDH G., MALMQUISTJ., 1959. Trypsin,invertase and amylase content of feces of germ-free rats. Proc. Soc. exp. Biol. Med., 102, 154-155,

BOYD G. S., EDWARDS H. M., 1967. Fat absorption by germ-free chicks. Poultry Sci., 46, 1481-1483.COATES M. E., HEWITT S., GOLOB P., 1970. A comparison of the effects of raw and heated soya-

bean meal in diets for germ-free and conventional chicks. Br. J. Nutr., 24, 213-225.COATES M. E., HEWITT D., SALTER D. N., 1973. Protein metabolism in the germ-free and conven-

tional chick, 291-295. In J. B. HENEGHAN, Germ-free research. Biological effect of gnotobioticenvironments. Acad. Press, New York and London.

COMBE E., 1971. Etude de I’azote endog6ne dans le contenu digestif des rats holox6niques et axeni- I.

ques a l’aide d’acides amines 1’C. C. R. Soc. Biol., 165, 289-295.COMBE E., ARNAL M., SACQUET E., 1967. Etude a I’aide d’acides amines marqués de I’origine des

composes azotes contenus dans le caecum des rats axdniques. C. R. Soc. Biol., 161, 1076-1079.COMBE E., DEMARNE Y., GUTGUEN L., IVOREC-SZYLIT O., MESLIN J. C., SACQUET E., 1976.

Some aspects of the relationship between gastrointestinal flora and host nutrition. VVId. Rev.Nutr. Diet., 24, 1-57.

COMBE E., GORDON H. A., 1969. Bidirectional permeation of water and amino acids in the cecumof germ-free rats, 169-170. In E. MIRAND, Germ-free biology, Plenum Publ. Co., New York.

COMBE E., PENOT E., CHARLIER H., SACQUET E., 1965. Métabolisme du rat « germ-free ». Teneurdes contenus digestifs en certains composes azot6s, en sodium, en potassium. Teneurs dequelques tissus en acides nucléiques. Ann. Biol. anim. Bioch. Biophys., 5, 189-206.

CORRING T., MOREAU C., DUCLUZEAU R., 1979. Comparative apparent digestibility of casein inholoxenic, axenic and Clostridium bifermanions monoassociated rats. Am. J. clin. Nutr., 32,1231-1237.

DAH LQVI5T A., 1963. Rat intestinal dextranase. Localization and relation to the other carbohydrasesof the digestive tract. Biochem. J., 86, 72-76.

DAHLQVIST A., BORGSTRdM B., 1961. Digestion and absorption of disaccharides in man. Biochem.’

J., 81, 411-418.

DAHLQVIST A., BULL B., GUSTAFSSON B. E.,1965. Rat intestinal 6-bromo-2-naphtyl glycosidase anddisaccharidase activities. Enzymatic properties and distribution in the digestive tract ofconventional and germ-free animals. Arch. Biochem. Biophys., 109, 150-158.

DAHLQVIST A., THOMSON D. L., 1964. The digestion and absorption of lactose by the intact rat.Acta physiol. scand., 61, 20-33.

DANIELSSON H., EINARSSON K., JOHANSSON G., 1967. Effect of biliary drainage on individualreaction in the conversion of cholesterol to taurocholic acid. Eur. J. Biochem., 2, 44-49.

DEMARNE Y., SACQUET E., FLANZY J., GARNIER H., FRANÇOIS A. C., 1970. Utilisation digestiveapparente des acides gras chez le rat Qx6nique et le rat holox6nique. Ann. Biol. anim. Bioch.Biophys.,10, 369-384.

DEMARNE Y., SACQUET E., FLANZY Y., GARNIER H., 1972. Influences cumul6es de la esecectomie

et de 1’6tat ox6nique sur I’utilisation digestive apparente de la ration et des acides gras chez lerat. Ann. Biol. anim. Bioch. Biophys., 12, 139-148.

DEMARNE Y., SACQUET E., GARNIER H., 1972. La flore gastrointestinale et la digestion des matiè-res grasses chez le monogastrique. Ann. Biol. anim. Bioch. Biophys., 12, 509-524.

DUCLUZEAU R., RAIBAUD P., 1975. La flore microbienne du tube digestif chez le monogastrique etson influence sur les m6tabolismes nutritionnels de I’h6te. Z. ErnBhrugswiss., 14, 133-144.

ECKNAUER R., RAFFLER H., 1978. Effect of starvation on small intestinal enzyme activity in germ-free rats. Digestion, 18, 45-55.

EINARSSON K., GUSTAFSSON J. A., GUSTAFSSON B. E., 1973. Differences between germ-free andconventional rats in liver microsomal metabolism of steroids. J. biol. Chem., 248, 3623-3630.

EYSSEN H. J., PARMENTIER G. G., MERTENS J., 1976. Sulfated bile acids in germ-free and conven-tional mice. Eur. J. Biochem., 66, 507-514.

GEEVER E. F., DAFT F. S., LEVENSON S. M., 1965. Pancreatic atrophy and fibrosis in the germ-freerat on a chemically defined diet. Fed. Proc., 24, 246 (abstr.).

GENELL S., GUSTAFSSON B. E., OHLSSON K., 1976. Quantitation of active pancreatic endopepti-dases in intestinal contents of germ-free and conventional rats. Scand. J. Gastroenterology, 11,757-762.

GLIMSTEDT G., 1936. Bakterienfreie Meerschweinchen-Aufzucht Lebensfdhigkeit und Wachstum,nebs untersuchungen uber das Lyphatische Gewebe. Acta path. microbial. scand., suppl., 30,1-295.

GORDON H. A., BRUCKNER-KARDOSS E., 1961. Effect of normal flora on intestinal surface area.

Am. J. Physiol., 201, 175-178.GORDON H. A., BRUCKNER-KARDOSS E., STALEY T. E., WAGNER M., WOSTMANN B. S., 1966.

Characteristics of the germ-free rat. Acta anat., 64, 367-389.GORDON H. A., WOSTMANN B. S., 1960. Morphological studies on the germ-free albino rat. Anat.

Rec., 137, 65-70.

GRAY G. M., COOPER H. L., 1971. Protein digestion and absorption. Gastroenterology, 61, 535-544.GUENET J. L., SACQUET E., GUENEAU G., MESLIN J. C., 1970. Action de la microflore totale du rat

sur I’activit6 mitotique des cryptes de Lieberkuhn. C. R. Acad. Sci. Paris, s6r. D, 270, 3087-3090.HEIZER W. D., KERLEY R. L., ISSELBACHER K. J., 1972. Intestinal peptide hydrolases differences

between brush-border and cytoplasmic enzymes. Biochim. biophys. Acta, 264, 450-461.HENEGHAN J. B., 1963. Influence of microbial flora on xylose absorption in rats and mice. Am.

J. Physiol., 205, 417-420.HERSKOVIC T., KATZ J., FLOCH M. H., SPENCER R. P., SPIRO H. M.,1967. Small intestinal absorp-

tion and morphology in germ-free, monocontaminated, and conventionalized mice. Gastro-enterotogy, 52, 1136.

HOFMANN A. F., MOSBACH E. H., SWEELEY C. C., 1964. Bile acid composition from germ-freerabbits. Biochim. biophys. Acta, 176, 204-207.

IVOREC-SZYLIT 0., RAIBAUD P., SCHELLENBERG P., 1973. The breakdown of starch at differentlevels in the digestive tract of the axenic, gnotoxenic and holoxenic chicken, 225-231. InJ. B. HENEGHAN, Germ-free research. Acad. Press, New York and London.

JERVIS H. R., BIGGERS D. C., 1964. Mucosal enzymes in the cecum of conventional and germ-freemice. Anat. Rec., 148, 591-595.

KAWAI Y., MOROTOMI M., 1978. Intestinal enzyme activities in germ-free, conventional and gnoto-biotic rats associated with indigenous microorganisms. Infect. lmmun., 19, 771-778.

KELLOGG T. F., 1971. Microbiological aspects of enterohepatic neutral sterol and bile acid metabo-lism. Fed. Proc., 30,1808-1814.

KELLOGG T. F., WOSTMANN B. S.,1969a. Fecal neutral steroids and bile acids from germ-free rats.J. Lipid Res., 10, 495-503.

KELLOGG T. F., WOSTMANN B. S., 19696. The response of germ-free rats to dietary cholesterol,293-295. In E. MIRAND, Germ-free biology. Plenum Publ. Co., New York.

KWONG E., GRUBER C., KAN D., BARNES R. H., LEVENSON S. M., 1971. Effect of soyabean tryp-sin inhibitor upon pancreatic cystine synthesis in germ-free rats. Fed. Proc., 30, 641 (abstr.).

LARNER J., GILLEPSIE R. E., 1957. Gastrointestinal digestion of starch. 111. Intestinal carbohydraseactivities in germ-free and non-germ-free animals. J. biol. Chem., 225, 278-285.

LEE H. W., MOINUDDIN J. F.,1958. Effects of oral tetracycline on diarrhea and on the dry weights ofcleaned organs in rats fed two different diets. Am. J. Physiol., 192, 421-426.

LEPKOVSKY S., WAGNER M., FURUTA F., OZONE K., KOIKE T.,1964. The proteases, amylase andlipase of the intestinal contents of germ-free and conventional chickens. Poultry Sci., 43, 722-726.

LEPKOVSKY S., FURUTA F., OZONE K., KOIKE T., 1966. The proteases, amylase and lipase of thepancreas and intestinal contents of germ-free and conventional rats. Br. J. Nutr., 20, 257-261.

LESHER S., WALBURG Jr. H. E., SACHER Jr G. A., 1964. Generation cycle on the duodenum cryptcells of germ-free and conventional mice. Nature, 202, 884-885.

LEVENSON S. M., GRUBER C., KAN D., 1969. Similarity in passage rates of plasma proteins into thegut of germ-free and conventionalized rats. J. Nutr., 98, 99-104.

LOESCHE W. J.,1968. Protein and carbohydrate composition of cecal contents of gnotobiotic rats andmice. Proc. Soc. exp. Biol. Med., 128, 195-199.

LOESCHKE K., GORDON H. A., 1970. Water movement across the cecal wall of the germ-free rat.

Proc. Soc. exp. Biol. Med., 133, 1217-1222.LUCKEY T. D., 1963. Germ-free life and gnotobiology, Acad. Press, New York, London, p. 252.

MALIS F., FRIC P., LOJDA Z., STEPANKOVA R., KRUML J., 1974. Verdanungsenzyme des Dunn-darms und Pankreas bein Keimfreien Tieren. Dent. Zeit. Verdon. Stofj’weckselkr., 34, 225-228.

MALIS F., FRIC P., STEPANKOVA R., KRUML J., 1976. Trypsin and chymotrypsin activity of the intes-tinal content in germ-free, monoassociated and conventional rabbits. Physiol. Bohemoslov, 25,71-74.

MASSON M. J., 1954. Microscopic studies of the gut flora of hen with special reference to the break-down of starches. Proc. 10th Wld Poultry Congr., 105-141.

MAYLIE M. F., CHARLES M., GACHE C., DESNUELLE P., 1971. Isolation and partial identification ofa pancreatic colipase. Biochim. biophys. Acta, 229, 286-289.

MESLIN J. C., 1971. Action de la microflore totale du rat sur 1’6pith6lium intestinal. Estimation de lasurface absorbante. Ann. Biol. anim. Bioch. Biophys., 11, 334-335.

MESLIN J. C., SACQUET E., GUENET J. L., 1973. Action de la flore bact6rienne sur la morphologieet la surface de la muqueuse de l’intestin grele du rat. Ann. Biol. anim. Bioch. Biophys., 13, 203-214.

MESLIN J. C., SACQUET E., RAIBAUD P., 1974. Action d’une flore microbienne qui ne ddconjuguepas les sels biliaires sur la morphologie et le renouvellement cellulaire de la muqueuse del’intestin gr6le du rat. Ann. Biol. anim. Bioch. Biophys., 14, 709-720.

MEYNELL G. G., 1963. Antibacterial mechanism of the mouse gut. II. The role of Eh and volatile

fatty acids in the normal gut. Br. J. exp. Path., 44, 209-219.PHILIPS B. P., WOLFE P. A., GORDON H. A., 1959. Studies on rearing the germ-free guinea pig.

Ann. N. Y. Acad. Sci., 78,183-207.PLEASANTS J. R., 1959. Rearing germ-free caesarian-born rats, mice and rabbits through weaning.

Ann. N. Y. Acad. Sci., 78,116-126.REDDY B. S., 1971. Calcium and magnesium absorption. Role of intestinal microflora. Fed. Proc. Fed.

am. Soc. exp. Biol., 30,1815-1821.REDDY B. S., PLEASANTS J. R., WOSTMANN B. S.,1968. Effect of dietary carbohydrates on intestinal

disaccharidases in germ-free and conventional rats. J. Nutr., 95, 413-419.REDDY B. S., PLEASANTS J. R., WOSTMANN B. S., 1969. Pancreatic enzymes in germ-free and

conventional rats fed chemically defined, water-soluble diet free from natural substrates. J.Nutr., 97, 327-334.

REDDY B. S., WOSTMANN B. S., 1966. Intestinal disaccharidase activities in the growing germ-freeand conventional rats. Arch. Biochem. Biophys., 113, 609-616.

RERAT A., 1978. Digestion and absorption of carbohydrates and nitrogenous matters in the hindgutof the omnivorous non-ruminant animal. J. anim. Sci., 46, 1808-1837.

RIEDEL G., SCHARRER E., LOSCH U., 1972. Resorptionstudien an Keimfreien Kuken. I. Aminosco-

renresorption. Zbl. Vet. Med., 19, 563-568.RIOTTOT M., SACQUET E., VILA J. P., LEPRINCE C., 1980. Relationship between small intestine

transit and bile acid metabolism in axenic and holoxenic rats fed different diets. Reprod. Nutr.Develop., 20, 163-171.

ROLLS B. A., TURVEY A., COATES M. E., 1978. The influence of the gut microflora and of dietaryfibre on epithelial cell migration in the chick intestine. Br. J. Nutr., 39, 91-98.

ROZE C., SACQUET E., CHARIOT J., DUBRASQUET M., ACCARY J. P., 1977. Quelques caractéris-tiques des secretions pancriatique et gastrique chez le rat axénique. Forum de Gastroentérologie,Paris.

SACQUET E., 1971. Fiore microbienne du tractus digestif et nutrition chez Rattus norvegicus. These Doct.Sci., Paris.

SACQUET E., GARNIER H., RAIBAUD P., 1970. Etude de la vitesse du transit gastro-intestinal desspores d’une souche thermophile stricte de Bacillus subtilis chez le rat « holoxenique », le rat« ax6nique » et le rat « axenique » caecectomise. C. R. Soc. Biol., 164, 532-537.

SACQUET E., VAN HEIJENOORT Y., RIOTTOT M., LEPRINCE C., 1975. Action de la flore micro-bienne du tractus digestif sur le m6tabolisme des acides biliaires chez le rat. Biochim. biophys.Acta, 380, 52-65.

SACQUET E., MEJEAN C., LEPRINCE C., RIOTTOT M., RAIBAUD P., 1976. Action du regime ali-mentaire et de la flore microbienne du tractus digestif sur le pool intestinal et 1’excr6tion fdcaledes acides biliaires chez le rat : 6tude compar6e chez des rats ax6niques, gnotox6niques et

holoxeniques. Ann. Nutr. Alim., 30, 603-617.

SACQUET E., LEPRINCE C., RIOTTOT M., 1979. Effect of different modifications of a semi-syntheticdiet on bile acid metabolism in axenic and holoxenic rats. Ann. Biol. anim. Bioch. Biophys., 19,1677-1688.

SALTER D. N., COATES M. E., 1971. The influence of the microflora of the alimentary tract on pro-tein digestion in the chicks. Br. J. Nutr., 26, 55-69.

SAVAGE D. C., SCHAEDLER R. W., DUBOS R. J., 1967. Effects of bacteria on gastrointestinal histo-logy. Fed. Proc., 26, 803 (abstr.).

SCHAEDLER R. W.,1973. The relationship between the host and its intestinal flora. Proc. Nutr. Soc. 32,41-47.

SCHIFF E. R., SMALL N. C., DIETSCHY J. M., 1972. Characterization of the kinetics of the passiveand active transport mechanisms for bile acid absorption in the small intestine and colon of therat. J. clin. Invest., 51,1351-1362.

SEIJFFERS M. J., SEGAL H. L., MILLER L. L.,1963. Separation of pepsinogen I, pepsinogen II and pepsi-nogen III from human gastric mucosa. Am. J. Physiol., 20, 1106-1116.

SIDDONS R. C., 1969. Intestinal disaccharidase activities in the chick. Biochem. J., 112, 51-59.SIDDONS R. C., COATES M. E., 1972. The influence of the intestinal microflora on disaccharidase

activities in the chick. Br. J. Nutr., 27,101-112.SPRINZ H., KUNDEL D. W., DAMMIN G. J., HOROWITZ R. E., SCHNEIDER H.. FORMAL S. B.,1961.

The response of the germ-free guinea pig to oral bacteria challenge with Escherichia coli andShigella flexneri : with special reference to lymphatic tissue and the intestinal tract. Am. J. Path.,39, 681-695.

SZYLIT 0., 1971. Etude compcirde de la degradation de I’amidon dans les caeca du pouletaxenique etorthox6nique. C. R. Acad. Sci. Paris, s6r. D, 273, 1132-1135.

SZYLIT 0., 1973. Les voies m6taboliques de la d6gradation de I’amidon dans le jabot de Gallus gallus etleurs incidences nutritionnelles. These Doct. Sci., Paris.

THORBECKE G. J., GORDON H. A., WOSTMANN B. S., WAGNER M., REYNIERS J. A.,1957. Lym-phoid tissue and serum gamma globulin in young germ-free chickens. J. Infect. Dis., 101, 237-251.

UKAI M., TOMURA A., ITO M., 1976. Cholesterol synthesis in germ-free and conventional rats. J.Nutr., 106, 1175-1183.

WHITT D. D., DEMOSS R. D., 1975. Effect of microflora on the free amino acid distribution in various

regions of the mouse gastrointestinal tract. Appl. Microbial., 30, 609-615.WOSTMANN B. S.,1973. Intestinal bile acids and cholesterol absorption in the germ-free rat. J. Nutr.,

103,982-990.WOSTMANN B. S., BRUCKNER-KARDOSS E., 1959. Development of cecal distention in germ-free

baby rats. Am. J. Physiol., 197, 1345-1346.WOSTMANN B. S., BRUCKNER-KARDOSS E., KNIGHT Jr P. L., 1968. Cecal enlargement, cardiac

output, and 02 consumption in germ-free rats. Proc. Soc. exp. Biol. Med., 128, 137-141.WOSTMANN B. S., BRUCKNER-KARDOSS E., BEAVER M., CHANG L., MADSEN D., 1976. Effect

of dietary lactose at levels comparable to human consumption on cholesterol and bile acidmetabolism of conventional and germ-free rats. J. Nutr., 106, 1782-1790.

YAMANAKA M., IWAI H., SAITO M., YAMAUCHI C., NOMURA T., 1972. Influence of intestinalmicrobes on digestion and absorption of nutrients in diet and nitrogen retention in germ-free,gnotobiotic and conventional mice. I. Protein and fat digestion and nitrogen retention in germ-free and conventional mice. Jap. J. Zootech. Sci., 43, 272-283.

YOSHIDA T., PLEASANTS J. R., REDDY B. S., WOSTMANN B. S., 1968. Efficiency of digestion ingerm-free and conventional rabbits. Br. J. Nutr., 22, 723-737.